The ability of malignant gliomas to diffusely infiltrate the surrounding brain has rendered them incurable by surgical resection, and the spread of tumor into critical brain areas contributes to tumor recurrence and the eventual loss of life associated with these lesions. However, the intracellular mechanisms regulating glioma cell motility and invasion remain poorly understood, and this has hampered efforts to develop therapies that prevent glioma invasion. We have identified a novel intracellular pathway that controls the cytoskeletal organization, migration and invasion of malignant glioma cells, and that is dysregulated in glioblastoma. This pathway involves the cyclin-dependent kinase-associated phosphatase, KAP (which is aberrantly spliced in glioblastoma), Rho kinase and Cdc2 kinase. Importantly, small molecule inhibition of this pathway using a Cdc2 kinase inhibitor decreases the migration and invasion of CD133+ glioma-derived stem-like cells in vitro and in the mouse brain. Moreover, dysregulation of this pathway defines a subpopulation of human glioblastomas associated with poor patient outcome. This project will examine the downstream effector mechanisms of this KAP/ROCK/Cdc2 invasion pathway in human CD133+ glioblastoma- derived stem-like cells. In addition, we will examine the effects of small molecule inhibition of Cdc2 kinase on glioblastoma dispersal and overall survival in vivo using intracranial transplantation of human glioma-derived stem-like cells into the mouse brain. Several small molecule Cdc2 inhibitors are currently being used in human clinical trials for other purposes, and may thus have therapeutic potential for use in clinical trials to prevent malignant glioma dispersal.